Comment 16 for bug 1946149

We've got a serial console log from AWS Support through our Support team
(special thanks to Pedro Principeza and our former colleague Mark Thomas.)

The problem is definitely not the ext4/jbd2 patchset as suspected
(although it's unclear how reverting it caused the kernel to boot;
maybe build environment differences?)

Early in the kernel boot, before even trying to mount the rootfs,
there are blcked swapper tasks, and they just continue to happen.

(full log attached.)

```
         Starting Reboot...
...
[ 0.000000] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-4.15.0-1113-aws root=UUID=db937f23-4ed7-4c4b-8058-b23a860fae08 ro console=tty1 console=ttyS0 nvme_core.io_timeout=4294967295
...
[ 0.000000] gran_size: 64K chunk_size: 256M num_reg: 10 lose cover RAM: 737G
...
[ 2.742455] clocksource: Switched to clocksource tsc
[ 242.656089] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
[ 363.488083] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
[ 484.320066] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
[ 605.152061] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
[ 725.984054] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
[ 846.816051] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
[ 967.648055] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
[ 1088.480033] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
[ 1209.312036] INFO: task swapper/0:1 blocked for more than 120 seconds.
...
<end of log>
```